Drafted with AI assistance, edited by Yahya Hafez.
Loading post…
Monthly notes on comprehension, accessibility, and AI tutoring. No spam, no upsells, unsubscribe anytime.
Ask a room of adults what dyslexia is and most will say something about seeing letters backwards, or b and d swapping places, or words sliding around on the page. It is one of the most durable ideas in popular psychology and it is wrong. Dyslexia is not a vision problem. It is a language problem, specifically a problem with the sound structure of words, and it leaves a measurable fingerprint on the reading brain that looks the same in English, Italian, and German. Getting the science right matters, because the myth points people toward fixes that do not work and away from the ones that do.
The working definition most researchers and many education systems use comes from the International Dyslexia Association. Dyslexia is a specific learning disability, neurobiological in origin, marked by difficulty with accurate or fluent word recognition and by poor spelling and decoding. Those difficulties typically result from a deficit in the phonological component of language, and they are unexpected given the person's other cognitive abilities and the instruction they received (International Dyslexia Association definition (opens in new tab)).
Notice what is absent. Nothing about vision. Nothing about intelligence. Nothing about reversing letters.
The "backwards reading" idea has been discredited since the 1970s, yet a 2020 review of public dyslexia education modules found that only two of fourteen bothered to address it directly (Anderson et al., 2020 (opens in new tab)). Letter reversals do happen, but they happen in nearly all beginning readers, because the visual system treats mirror images as the same object until reading instruction trains it out. A child can be severely dyslexic and never reverse a letter. The reversals are a symptom of being new to print, not of dyslexia.
To read a word you have never seen, you have to map letters onto the speech sounds they stand for, hold those sounds, and blend them. That operation depends on phonological awareness, the ability to notice and manipulate the sound units inside spoken words. Dyslexic readers find this hard even with their eyes closed. They struggle to say "cat" without the first sound, to hold a string of nonsense syllables in mind, or to retrieve the name of a letter quickly. The letters on the page are seen perfectly well. The trouble starts when the brain tries to turn them into sound.
This is why decoding breaks down. A skilled reader meets an unfamiliar word, sounds it out once or twice, and stores it as a unit. A dyslexic reader has to repeat that effortful sounding-out many more times before the word becomes automatic, so word recognition stays slow and effortful long after peers have moved on to fluent reading. Comprehension suffers not because the reader cannot think, but because so much working memory is consumed by decoding that little is left for meaning.
Researchers have not always agreed that phonology is the whole story. The magnocellular theory proposed that dyslexia stems from a low-level sensory deficit in the fast visual and auditory pathways, which would make the phonological problems a downstream effect. A cerebellar theory blamed motor and automaticity systems. Each had supporting studies, usually showing that some dyslexic participants performed poorly on some sensory or motor task.
The most careful attempt to adjudicate was a 2003 multiple case study in the journal Brain. Franck Ramus and colleagues gave sixteen dyslexic university students and sixteen controls a full battery of phonological, auditory, visual, and cerebellar tests, then looked at each person individually instead of averaging across the group (Ramus et al., 2003, open preprint (opens in new tab)). All sixteen dyslexic adults had a phonological deficit. Ten also had an auditory deficit, four a motor deficit, and only two a visual magnocellular deficit. The authors concluded that a phonological deficit can occur on its own and is sufficient to cause a reading impairment, while sensory and motor problems appear in a subset and may aggravate the phonological one without causing it.
That pattern has held up. Sensory differences are real for some dyslexic people, which is one reason visual accommodations help, but the thing every dyslexic reader shares is the phonological core.
In 1998, Sally and Bennett Shaywitz and colleagues at Yale scanned 29 dyslexic and 32 typical adult readers with functional MRI while they performed tasks that demanded progressively more phonological analysis (Shaywitz et al., 1998 (opens in new tab)). Dyslexic readers showed underactivation in posterior left-hemisphere regions, including the angular gyrus and Wernicke's area, and overactivation in the inferior frontal gyrus. The posterior regions are where the reading brain normally wires print to sound; the frontal overactivation looks like a compensatory effort, the brain working harder through a less efficient route.
Two decades of follow-up studies have refined the map. A 2012 review by Fabio Richlan argued that the evidence points to dysfunction across a left-hemisphere reading network rather than a single temporo-parietal spot, with the left occipito-temporal region, home of the so-called visual word form area, as the most consistent site of underactivation (Richlan, 2012 (opens in new tab)). A 2016 meta-analysis of 28 imaging studies then asked whether this holds across languages. It compared studies in English, a deep orthography where spelling and sound match loosely, with studies in Dutch, German, Italian, and Swedish, where they match tightly. The left occipito-temporal underactivation appeared in both (Martin, Kronbichler & Richlan, 2016 (opens in new tab)). Dyslexia has a shared neural signature whatever language you read in.
One more scan result deserves a place in the myth-busting column. A 2011 study of 131 children compared poor readers with high IQ scores against poor readers with low IQ scores (Tanaka et al., 2011 (opens in new tab)). Both groups showed the same reduced activation in the same left parietotemporal and occipitotemporal regions. Dyslexia looks the same in the brain regardless of general intelligence, which is why the old practice of diagnosing it only when reading lags behind IQ has been abandoned.
Since the problem is phonological and neurological, the fixes that work are the ones that either strengthen the sound-to-letter mapping or lower the cost of decoding. Nothing on this list involves vision therapy or colored lenses.
The person who sees letters backwards does not exist. The person whose brain wires print to sound through a slower, more effortful route does, in roughly one in ten of us, and that reader does far better once the page, and the people around them, stop solving the wrong problem.
